Computers & Electronics

A randomized, controlled trial of emotional disclosure in rheumatoid arthritis: Can clinician assistance enhance the effects?

A randomized, controlled trial of emotional disclosure in rheumatoid arthritis: Can clinician assistance enhance the effects?
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  A Randomized, Controlled Trial of Emotional Disclosure inRheumatoid Arthritis: Can Clinician Assistance Enhance theEffects? Francis J. Keefe 1, Timothy Anderson 2, Mark Lumley 3, David Caldwell 1, David Stainbrook 2, Daphne Mckee 1, Sandra J. Waters 1, Mark Connelly 1, Glenn Affleck 4, Mary Susan Pope 1, Marianne Weiss 2, Paul A. Riordan 1, and Brian D. Uhlin 21  Duke University Medical Center  2 Ohio University 3 Wayne State University 4 University of Connecticut  Abstract Emotional disclosure by writing or talking about stressful life experiences improves health status innon-clinical populations, but its success in clinical populations, particularly rheumatoid arthritis(RA), has been mixed. In this randomized, controlled trial, we attempted to increase the efficacy of emotional disclosure by having a trained clinician help patients emotionally disclose and processstressful experiences. We randomized 98 adults with RA to one of four conditions: a) private verbalemotional disclosure; b) clinician-assisted verbal emotional disclosure; c) arthritis informationcontrol (all of which engaged in four, 30-minute laboratory sessions); or d) no-treatment, standardcare only control group. Outcome measures (pain, disability, affect, stress) were assessed at baseline,2 months following treatment (2-month follow-up), and at 5-month, and 15-month follow-ups. Amanipulation check demonstrated that, as expected, both types of emotional disclosure led toimmediate (post-session) increases in negative affect compared with arthritis information. Outcomeanalyses at all three follow-ups revealed no clear pattern of effects for either clinician-assisted orprivate emotional disclosure compared with the two control groups. There were some benefits interms of a reduction in pain behavior with private disclosure versus clinician-assisted disclosure atthe 2 month follow-up, but no other significant between group differences. We conclude that verbalemotional disclosure about stressful experiences, whether conducted privately or assisted by aclinician, has little or no benefit for people with RA. Introduction Studies suggest that written or verbal emotional disclosure about stressful events by writingor talking leads to temporary increases in negative affect, followed by improved health duringsubsequent months, at least with healthy populations (Smyth 1998). Recent studies have testedthis technique in clinical samples, but with weaker results (Frisina et al., 2004). Emotional Address correspondence to: Francis J. Keefe, Duke Pain Prevention and Treatment Research Program, Box 3159 Duke University MedicalCenter, Durham, North Carolina, USA, telephone: 919-416-3401, fax: 919-416-3458, email: Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customerswe are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resultingproof before it is published in its final citable form. Please note that during the production process errors may be discovered which couldaffect the content, and all legal disclaimers that apply to the journal pertain. NIH Public Access Author Manuscript Pain . Author manuscript; available in PMC 2009 July 1. Published in final edited form as: Pain . 2008 July ; 137(1): 164172. N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t    disclosure may help people with rheumatoid arthritis (RA) because they experience elevatedstress (Koehler, 1985; Rimon & Laakso, 1985), are reactive to stress (Thomason et al., 1992;Zautra et al., 1994; Potter & Zautra, 1997) and tend to inhibit expressing negative emotions(Fernandez et al., 1989). To date, four published randomized trials have tested disclosure inRA, with mixed results. Smyth et al. (1999) found that RA patients who wrote about stress for3 days in the laboratory had better physician ratings of disease 4 months later than patientsengaged in neutral writing. Kelley et al. (1997) demonstrated that RA patients randomized totalk about stress into a tape recorder at home for 4 days had better self-reported affective andphysical functioning—but not pain or joint condition—3 months later than did controls.Wetherell et al. (2005) found that RA patients who disclosed at home (either writing orspeaking) had better mood and less disease activity than controls 10 weeks later, but theseeffects were due to unexpected worsening among controls, rather than improvements amongthe disclosure group. Finally, Broderick et al. (2004) found that 3 days of at-home writtendisclosure writing had no effects at 6-month follow-up, although an enhanced disclosurecondition showed equivocal benefits, confounded by pre-treatment group differences.Research is needed on methods to enhance disclosure’s effects. Studies suggest thatparticipants may benefit from instructions that help them identify and express feelings (Sloanet al. 2007), remain on topic over disclosure days (Sloan et al. 2005), make a coherent narrative(Smyth et al. 2001) and explore the meaning of the experience (Pennebaker et al. 1997; Gidronet al. 2002; Ullrich & Lutgendorf 2002; Broderick et al. 2004). Studies of emotional disclosurevia the internet using individualized therapist feedback have shown substantial effects (Langeet al., 2001; 2003), and disclosure to a therapist appears to create as much improvements asprivate disclosure, but with less negative affect (Murray et al. 1989; Donnelly & Murray1991; Segal & Murray 1994; Anderson et al. 2004).We hypothesized that clinician assistance would increase the benefits of disclosure because aclinician can help patients identify and verbalize their feelings, remain on topic, and reflect onthe meaning of their experience. We compared clinician-assisted disclosure to private verbaldisclosure and a neutral arthritis information control condition and a standard care group. Wetested verbal rather than written disclosure because verbal disclosure can be effective for RA(Kelley et al., 1997) and disclosure to and assistance by a clinician is typically verbal. Wetested not only immediate mood effects of disclosure but also effects at 15 month follow-up,which is much longer than prior studies of disclosure in RA. Methods Subjects Patients with RA were recruited from clinics affiliated with the Ohio University College of Osteopathic Medicine or Duke University Medical School. Recruitment occurred between Mayof 2000 and December of 2003. All patients were given a history and physical examination byone of the study rheumatologists and included only if they met 1987 American College of Rheumatology criteria for the diagnosis of RA. Patients were excluded if they had other organicdisease that would significantly affect function (e.g., COPD) or rheumatic disorders other thanRA. Patients with severe personality disorders (e.g., borderline personality disorder), substanceabuse problems, or who were involved in current psychiatric treatment were excluded. Procedure After providing consent, patients were screened to determine if they met eligibility criteria.They next completed a baseline assessment to assess pain, physical disability, psychologicaldisability, stress, and affect and then were randomly assigned to one of four conditions: a)private emotional disclosure, b) clinician-assisted emotional-disclosure, c) arthritis Keefe et al.Page 2 Pain . Author manuscript; available in PMC 2009 July 1. N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t    information, or d) standard care. Randomization was done by concealment with assignmentsin sealed envelopes and investigators and patients unaware of treatment condition until the dateof randomization. All patients, except those in the standard care condition, attended four, 30-min sessions within a 3 week time interval in the research clinic (mean number of days fromstart to completion of treatment=21.3 days). All treatment sessions were audiotaped andparticipants were instructed that audiotapes would be listened to by members of the researchstudy staff. To ensure confidentiality, session audiotapes were marked with an identificationnumber rather than participants' names. Participants in the three intervention groups rated theirnegative affect before and after each session. Follow-up assessments were conducted 2 monthsafter the end of the treatment period (2-month follow-up) and 5, and 15 months after thetreatment period. Figure 1 is a CONSORT diagram that provides an overview of the studydesign and information on numbers of patients evaluated at each time point. Intervention ConditionsPrivate emotional disclosure— Participants in this condition spent 30 minutes during eachof four sessions, alone in a private room in the clinic, talking into an audiotape recorder.Instructions were similar to those used in standard disclosure studies (Pennebaker 1995), butmodified for a verbal, tape-recorded format as done by Kelley et al. (1997). Participants wereinstructed to identify an unresolved stressful experience in their lives, and they were given cuesto facilitate this (e.g., an experience that is uncomfortable to talk about or remember, that makesone feel anxious or upset, that is avoided when possible, or that crosses the mind frequently).They were instructed to talk about this experience, including both the facts and their deepestfeelings about it and to label their feelings. They also were instructed that they could explorehow the experience is related to how they are dealing with rheumatoid arthritis, and they coulddiscuss how the experience relates to their childhood, family, parents, children, and friends.They were encouraged to “work on and resolve one stressful experience at a time, even if itmeans talking about the same experience over several days. However, if you find that you haveworked it out or feel better about it, you should go on and talk about another stressful topic.” Clinician-assisted emotional disclosure— The clinician-assisted emotional disclosureprotocol was designed to retain the structural and practical advantages of a standard 4-sessionprivate disclosure protocol, but to enhance the patient’s experience through the use of emotionalfocusing techniques in the form of a brief, nurse-administered protocol. This protocol wasdeveloped for this study by integrating the findings of studies that suggest the benefits of emotional exploration, narrative structure, and meaning development (Smyth et al. 2001;Gidron et al. 2002; Broderick et al. 2004; Sloan et al. 2005; Sloan et al. 2007) with specifictechniques taken from Emotion-Focused Therapy (Greenberg et al., 1993). Participants in thiscondition met with a nurse for four, 30-minute individual sessions and were provided the sameinstructions as the private disclosure group. The nurse was trained to create and maintain arelationship by using techniques of active listening, reflection, and avoiding negative judgments to help the patient feel a sense of positive regard and acceptance. Against thisbackdrop of a positive relationship, the nurse used two techniques to facilitate emotionaldisclosure: a) discovering and elaborating feelings, and b) emotional focusing. The firstcomponent involves assisting the patient to identify problematic or stressful experiences,discuss the experiences surrounding the problematic event and to search for relevant internaland external details, to track the personal meanings that they assign to the causes of the eventand their own reactions, and to engage in a broader re-examination of the event. Emotionalfocusing is embedded within the process of discovering and elaborating narratives. Thetechnique of experiential focusing (Greenberg, et al, 1993) encouraged patients to try to focussolely on feelings about the problem event while avoiding thinking about interfering thoughts.Patients were helped to identify as many different aspects of the emotional experience aspossible and develop verbal labels to describe their experience. Keefe et al.Page 3 Pain . Author manuscript; available in PMC 2009 July 1. N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t    Arthritis information control condition— Participants in this control condition met witha nurse for four, 30-minute individual sessions and were provided basic information aboutrheumatoid arthritis. The arthritis information sessions used a lecture-discussion format basedon the early work of Lsrc (1982) and sessions were supplemented by charts and photographs.The protocol provided information about the nature of rheumatoid arthritis disease, treatmentof rheumatoid arthritis, exercise, and methods for joint protection and maintaining mobilityand function. This condition was designed to serve as a control for time, attention, andinteraction with a nurse. Standard Care Control Condition— The standard care condition was designed to serveas a routine treatment control. These patients did not attend individual sessions for emotionaldisclosure or arthritis information, but they did complete all measures at time intervalscorresponding to the baseline and the three follow-up assessments. This condition served as acontrol for the passage of time. Nurse Training— Treatment sessions for the clinician-assisted emotional disclosure andarthritis information sessions were conducted by registered nurses each of whom had over 20years of experience in patient education and psychosocial interventions. To control for nurseeffects, the same nurses conducted both the clinician-assisted disclosure and the arthritisinformation sessions. Nurses were taught the clinician-assisted emotional disclosure protocoland arthritis information protocol in a 4-day didactic and experiential course (2 days each). Forclinician-assisted emotional disclosure, nurses were provided with a written protocol that wasreviewed during didactic sessions. During these didactic sessions, transcripts of treatmentsessions were used to illustrate training issues (e.g., identifying generic narratives and ways to“unfold” the story by encouraging the narrator to identifying emotional “markers”). Videotapesof model cases were also used to illustrate key points. Nurses then role-played commonscenarios that could arise in treatment. After the training seminar, each nurse was assigned twopractice cases (normal volunteers) in which they utilized the 4-session protocol. Training inthe arthritis information protocol was similar in that nurses practiced common scenarios likelyto arise in the sessions. They also practiced the protocol with two practice cases (normalvolunteers). All of the actual clinician-assisted emotional disclosure sessions and arthritisinformation sessions were audio-taped, and tapes were reviewed in weekly supervisionsessions with a supervisor in which feedback on adherence to the protocol was provided. MeasuresImmediate affect reactions to sessions— Patients in the three active conditionscompleted the Positive Affect Negative Affect Schedule (PANAS; Watson et al., 1988)immediately before and after each session. Items were rated on a 1 – 5 scale and summed. Wescored the 10-item negative affect (NA) scale and analyzed change scores (post-session minuspre-session). Internal consistency of NA for this sample was examined using Cronbach's alphaand found to be high ( α  = 0.929). Outcome Measures: We assessed pain, physical disability, psychological disability, painbehavior, stress, and affect (positive and negative) at baseline as well as at 1-, 3-, and 12-monthfollow-ups. We included measures of physical disability, psychological disability, and affectbecause prior studies of emotional disclosure in RA patients had shown that disclosure canproduce significant improvements in one or more of these outcomes (Smyth, et al., 1999;Kelley, et al. 1997; Wetherell et al., 2005; Broderick, et al., 2004). We included measures of pain, pain behavior, and stress because we believed that clinician-assisted emotional disclosuremight enhance the effects of prior disclosure enough to produce improvements in theseimportant RA outcomes. Keefe et al.Page 4 Pain . Author manuscript; available in PMC 2009 July 1. N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t    Outcome measures of pain, physical disability, and psychological disability— The Arthritis Impact Measurement Scale (AIMS; Meenan et al., 1982; Meenan et al., 1991)was used to measure pain, physical disability and psychological disability over the past month.Internal consistency of each of the AIMS measures was found to be high (pain: α = 0.824;physical disability: α = 0.807; psychological disability: α = 0.925). Research has supported thereliability of the AIMS and found it is valid when used with different types of arthritis, with arange of social and demographic groupings, and in different clinical settings (Meenan et al.,1982). Daily Stress— Levels of perceived stress were assessed using the Daily Stress Inventory(DSI, Brantley et al., 1987). The DSI is a 58-item self-report measure that yields measures of both the frequency of stressors and severity of stressors over the past day. Prior research hasindicated that the DSI has good test-retest reliability. Over a 28 consecutive day period thetest-retest reliability for the frequency of stressors was 0.72 (Brantley et al, 1987). Researchindicates that the DSI has good validity when used with patients having RA in that the DSIfrequency of stressors is very highly correlated with the DSI severity of stressors measure(Parker et al., 1995). Thus, for this study, the frequency of stressors was used as the outcomevariable. Positive and negative affect— The PANAS was used to assess affect and both positiveaffect (PA) and NA scores were calculated. Internal consistency for both scales was found tobe high ( α  = 0.917 for positive affect and α  = 0.899 for negative affect.). Pain behavior measure— Pain behavior was recorded using an RA version (McDaniel,Anderson, Bradley, et al., 1986) of an observation method developed in our lab (Keefe & Block,1982) which consists of a 10-minute session in which the patient engages in a standard, timedset of activities: sitting, walking, standing and reclining. Trained observers recorded theoccurrence of 7 categories of behavior (guarding, bracing, grimacing, sighing, rigidity, passiverubbing, and active rubbing). Two observers independently coded 32% of the pain behaviorsamples and the inter-observer reliability was 85% (percentage agreement). A composite score,total pain behavior, was calculated based on the frequency of each pain behavior. Previousresearch has supported the reliability and validity of the total pain behavior measure (McDaniel,et al. 1986; Anderson, Bradley, McDaniel, et al. 1987;Anderson, Keefe, Bradley, et al.,1988).Content Coding of Disclosure Sessions. For both the private disclosure and clinician assisteddisclosure conditions, each of the events discussed in the session was content coded. Eventswere coded in the following categories: (a) RA experience; (b) non-RA physical problems; (c)interpersonal conflict; (d) general life stressors and trauma; (e) loss through death. Thecategories were not mutually exclusive. Prior to content coding, two coders who were notinvolved in the content coding, followed Angus, Levitt, and Hardtke’s (1999) method foridentifying shifts in narrative structure, referred to as sequences.  Sessions had a mean of 17.6.(SD = 6.3) sequences, which ranged from 2 to 40 sequences per session. Each content categorywas calculated as a percentage of sequences in which the event content had been identifiedwithin the session. Results Participant Characteristics Ninety-eight participants (73.5% women; 90.8% Caucasian) were randomized into the study.On average, the participants were 56 years old ( SD  = 10.59) and reported having hadrheumatoid arthritis for 14 years. In terms of education level, 29% had a high school degree Keefe et al.Page 5 Pain . Author manuscript; available in PMC 2009 July 1. N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  N I  H -P A A  u t  h  or M an u s  c r i   p t  
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!